User:Prioryman/Gibraltarpedia, monetisation and myths

(Author's note: I founded WikiProject Gibraltar in 2007 and am also a member of the Gibraltarpedia project. This essay represents my own personal views and shouldn't be seen as representative of the views of other Gibraltarpedia contributors.)

The Gibraltarpedia project is a collaborative effort aimed at improving Wikipedia's coverage, in multiple languages, of Gibraltar and the surrounding region in Spain and northwest Africa. It's supported by the Government of Gibraltar and was given a high-profile launch at Wikimania in July 2012. On 13 September 2012, a question posted on Wikipedia talk:Did you know sparked a controversy that led to accusations of paid editing, conflicts of interest and commercialisation of Wikipedia.

I'm going to leave aside the questions of conflicts of interest and paid editing (which, for the record, isn't happening; nobody is being paid a penny to write anything) and focus on something that hasn't been addressed in any depth so far. Does Gibraltarpedia actually deliver a commercial benefit? In other words, is it spamming Wikipedia to make profit for businesses in Gibraltar?

There is, of course, a wider question here: does any Wikipedia content have a commercial effect, pro or con? As far as I know, there's no evidence that it does. To the best of my knowledge no studies have found links between Wikipedia content and the conversion of sales. This doesn't mean that there definitely is no link; but if a link did exist, one would think that it would have been found and publicised by now, given Wikipedia's prominence and reach. It is possible that Wikipedia articles may have some marginal impact on corporate reputations but I doubt that you could quantify that in terms of sales.

Assessing Gibraltarpedia's impact
So, given this lack of hard evidence, is there any way we can assess the impact that Gibraltarpedia is having outside Wikipedia, and the possible scale of any commercial benefits that it might deliver? Fortunately, there is. A good starting point is to look at the 37 new articles listed at GLAM/GibraltarpediA and nominated for Did You Know? on the Main Page. It's immediately apparent from this list that there is not a lot of scope for conversion in the first place. The vast majority of the articles (21 in all) relate to buildings and places. 6 are historical articles, 3 are historical biographies and another 3 are biographies of living persons. 3 concern state or religious institutions (the Gibraltar Museum, the Royal Naval Hospital Gibraltar and the Synagogues of Gibraltar). Only 1 concerns a business, Profield Contractors.

The heavy focus on geographical articles does not seem to be planned but is probably a sign of the "low hanging fruit" effect – places are among the easiest and most obvious things to write encyclopedic content about. (Wikipedia itself developed in a similar fashion – in the early days much of the content consisted of geography stubs.) It also fits with the aim of Gibraltarpedia to add QR codes to places so that visitors can look up articles on them, though this is more likely coincidence rather than deliberate intention (so far at least). Needless to say, articles about places such as the Gibraltar War Memorial do not seem to be obvious candidates for commercialisation.

What evidence is there for articles produced as part of Gibraltarpedia, as opposed to Gibraltarpedia itself, being widely picked up outside Wikipedia? Actually, not much. Social media provides a useful insight into public interest in Gibraltarpedia's output. If Gibraltarpedia articles really are making waves, one would expect to see (1) lots of incoming links from social media outlets such as Twitter and (2) high numbers of page views of Gibraltarpedia DYK articles.

This isn't borne out by what we can see in practice. With only one exception (Georgina Cassar), so far the only outlets tweeting about Gibraltarpedia DYKs have been the @didyouknowwp and @gibraltarpedia Twitter feeds and a couple of Gibraltarpedia project members. The number of page views resulting from Gibraltarpedia articles appearing on DYK is quite low, ranging from 473 to 8,563. For comparison, my personal range over nearly 100 DYKs is from about 1,500 to 51,710; a project I conducted with support from Wikimedia UK to commemorate the RMS Titanic anniversary last April resulted in nearly 1 million page views from Main Page articles. Newness of articles also doesn't trigger massive amounts of initial page views, nor does it trigger automatic social media linking. The Titanic case illustrates the fact that with very few exceptions (Today's Featured Article being the obvious one), usage is driven primarily by external events – think Tom Cruise and Katie Holmes divorcing – rather than anything we do on Wikipedia.

Can Gibraltarpedia articles be monetised?
So the articles themselves are not attracting a huge amount of external attention. What about the links and sources used in them? Could that be a possible vehicle for conversions to sales? Let's consider Jews' Gate Cemetery, which got 1,401 views at DYK and was tweeted only once, by the DYK Twitter feed. If the aim is to provide a commercially useful article, the links would have to be capable of being monetised and converted to sales so that people can follow them through to a transaction.

There are 12 sources used on the article. 2 are government sources. 2 of these sources are pdfs. PDFs are not one click sales conversions and neither PDF file is a sales file. jewishvirtuallibrary.org is run by the America-Israeli Cooperative Enterprise, who do not appear to have a financial interest in promoting the businesses linked on the page. jewishgibraltar.com does offer tours of local synagogues but not very prominently, as the tour link is located at the very bottom of their page. The page in question is not a sales page and does not list costs. iajgsjewishcemeteryproject.org, websters-dictionary-online.com and jewishgen.org have no adverts. Three of the sources do have adverts: gibraltarinfo.gi, ecom.toyota-gib.com, and gibraltarinformation.com. Thus only a quarter of the linked pages have advertisements or are explicitly commercial sites. There is clearly little opportunity to convert Wikipedia traffic views from this article to sales, even before we take into account the low amount of traffic in the first place.

We know already from other GLAM projects that significant increases of traffic to the websites of collaborating organisations have not materialised. The British Museum, about a year into the GLAM/BM project, reported that although all articles on items in its collections had links to the BM website, the level of clickthroughs was low. This certainly matches my own experience that readers tend not to click through to reference links anyway. So whatever else Gibraltarpedia is going to produce, it's unlikely to be an effective way of promoting external websites.

Can Gibraltarpedia successfully promote Gibraltar?
What if the goal, as suggested by some of the comments from Gibraltarian sources quoted by media outlets, is to promote Gibraltar itself? It does seem likely that the Gibraltar Tourist Board is hoping to boost public awareness of the territory. A possible proxy for working out whether this is succeeding is to look at the statistics for the main Gibraltar article. Unfortunately the stats for the last year are interrupted by periods for which data is missing, but excluding months with incomplete data, the average number of monthly page views is about 121,000. In the last 30 days before 26 September – this of course includes the period of the controversy on Wikipedia - the total number of page views is 125,756, which is actually below the level of usage of several other months this year. Interestingly, there is little sign that the controversy itself has prompted a greater level of usage; after a brief peak of over 6,000 views on 19 September, usage has subsided to its usual average of about 4,000 page views a day.

We can see something similar on Twitter. There have been 64 Twitter links to the Gibraltar article. 33 of these were made over a year ago, while only 10 appear to have been made over the last 90 days. This very low volume does not suggest an increase in usage of Gibraltar as a result of Gibraltarpedia.

We can also consider the case of the Gibraltar Museum, one of the partners in Gibraltarpedia. http://stats.grok.se/en/latest/Gibraltar_Museum is their page views. http://www.alexa.com/siteinfo/gibmuseum.gi is the Gibraltar museum. http://www.quantcast.com/gibmuseum.gi on another site. The museum is not getting measurable traffic as a result of its exposure on Wikipedia. The stats for the latest 90 days on the Wikipedia article show a spike because of DYK with an overall downward trend in usage. So even a partner organisation isn't getting much direct benefit from Gibraltarpedia.

Gibraltarpedia's likely outcomes
What, then, are the likely outcomes of Gibraltarpedia? Media coverage is one obvious thing. The earlier Monmouthpedia project attracted a considerable amount of free publicity for the town of Monmouth. The Gibraltar Tourist Board is clearly hoping for something similar. Stories such as this one from the BBC are ideal from that point of view. (Forget the recent bad publicity; it was virtually all in the tech press or on Fox News, whose readers are never going to visit Gib anyway. There's no indication that it's had any effect on usage) However, I think the Tourist Board might be a bit over-optimistic about the likely scale of coverage. Monmouthpedia got a lot of coverage because it was genuinely novel. Gibraltarpedia is "more of the same but bigger" which is much less interesting from a media perspective. In any case, media coverage is transient and isolated stories don't have much effect on public opinion.

The QR codes being added to locations in Gibraltar may have some marginal effects locally, but this will be diluted by the fact that mobile Internet coverage in Gibraltar is expensive. (This is presumably why free wi-fi has been mooted. Unless that becomes a reality, I suspect the QR codes will not get much usage.)

The biggest and longest-lasting benefit is likely to be on Wikipedia itself, with the production of numerous new and improved articles on an important but underrepresented topic. That outcome is, I think, something that everyone can recognise as beneficial – if people are willing to look beyond innuendo and assumptions. Prioryman (talk) 22:50, 27 September 2012 (UTC)